Skip to main content
Think of tokens as the “units” of work for AI models. Larger prompts, more open files, and longer conversations use more tokens.

​
Why Tokens Matter

  • Tokens reflect the compute the AI needs to process your request
  • More context = more tokens
  • Managing context keeps costs lower and quality high

​
How Usage Works

  • Annual Membership ($33/year) grants access to Softgen’s builder platform, including integrations with GitHub, Supabase, and Vercel.
  • Pay‑as‑You‑Go: Add funds to your wallet; usage deducts transparently
  • No monthly bundles or rollover: Just keep your wallet funded to continue building

​
Tips for Token Efficiency

  1. Be specific: Clear, actionable requests use fewer tokens
  2. Manage context: Close unneeded files and summarize long threads
  3. Use the Context Monitor: Green is healthy up to ~150k tokens; trim above that
For deeper guidance on maintaining context health, see our Context‑Window & Thread Management guide.